California Sues E-Voting Vendor ES&S

Gustoman writes with news that the California Secretary of State has sued ES&S, a vendor of e-voting machines, for selling machines that were modifications of the model that has been certified. Apparently ES&S relocated two circuit boards, rerouted several internal cables, and changed some mounting bracket supports in their AutoMark A100 devices, named the modified version AutoMark A200, and sold 972 of them to five California counties. The changes sound somewhat trivial, but the certification contract specified that no "substitution or modification of the voting systems shall be made with respect to any component of the voting systems... until the secretary of state has been notified in writing and has determined that the proposed change or modification does not impair the accuracy and efficiency of the voting systems sufficient to require a reexamination and approval." The state is seeking a penalty of $10,000 per machine sold, plus the cost of the machines to the counties — almost $15 million in all.

Comments

Any hope?

[Opportunist]

Could they go bankrupt and we won't see any more voting machines?

Re:Any hope?

[Anonymous Coward]

Even better, perhaps this could scare companies away from making the machines even if they don't go bankrupt. :)

Re:Any hope?

[deniable]

I was thinking the same thing. The problem is you can raise the bar high enough that corruption becomes viable for both sides. This case will hopefully make the chuckle-heads behind the voting machines realize that they need to be building bulletproof systems and not barely good-enough consumer goods. Think embedded system rather than an MS Access 'solution.'

Re:Any hope?

[Opportunist]

That's pretty much the logic conclusion I drew from it.

I'm quite aware that this ain't the only company making voting machines. But if companies notice that they'll be held responsible and monitored tightly for the machines they make, they might consider it not worth the risk.

Re:Any hope?

[Slashidiot]

For me, it has always been surprising that you americans have all these problems with voting machines. Voting is a simple enough proccess, why would anybody need a machine to do it?

I think it's always better to do it the traditional way, you go there with your ballot and put it on a clear box, after somebody has checked that you are who you say you are, and that you are supposed to vote. Painfully simple, completely fool-proof. It takes a bit more people to do it, but it's just as being in a jury.

I see absolutely no downside about doing it the traditional way. Is there any reason to do it with machines in america, or you do it that way just because it's cooler?

Re:Any hope?

[hazem]

The answer to your question is in your post.

It's true that voting is simple process. Rigging a vote, however, is not as simple; and printing ballots does not have as high a profit margin as selling a voting machine.

The reasons for the machines and the reasons for the non-traditional way are: to make more money for friends (and campaign contributors) of politicians and to facilitate getting the desired (and paid for) result from an election.

It has nothing to do with the intelligence or lack thereof in the American voting populace.

Re:Any hope?

[batkiwi]

That "system" also makes:
-corruption/subversion trivial
-counting error prone
-counting OFTEN up to human judgement, which is never unbiased (hanging chad, etc)

Great one, there.

Re:Any hope?

[houghi]

The Americans and many other countries face a difficult choice here. On the one side we have instand reports on the outcome of the voting even before the voting is completely over. On the other side we have the democratic process.

One side is money, the other side is the people.

Darn, which one could be more importand?

Re:Any hope?

[Atario]

You'll have to forgive our desperate, gun-shy search for an alternative to paper ballots; you see, back in 2000, we had this problem with them, and, well, we wound up kinda screwed...

Re:Any hope?

[QuantumG]

counting OFTEN up to human judgement, which is never unbiased (hanging chad, etc) hanging chad? Did you miss the whole freakin' point of the discussion or what?

The rest of the democratic world doesn't have chads. We don't punch cards.

Re:Any hope?

[QuantumG]

Dude, there seems to be something wrong with your keyboard. For some reason you're typing 'd' instead of 't' at the end of sentences. Are you trying to learn dvorak or something?

Re:Any hope?

[KrizDog]

I see a few advantages to computer based voting systems that could greatly improve the transparency and accessibility of the process.

  The most obvious advantage is the privacy it offers visually impared people who would otherwise have to depend on others to follow they wishes. (Kinda uniformed about this so don't kill me if I'm wrong and they have simple usable brail forms)

A post above mentioned doing away with the secrecy aspect in the hope that they can audit their vote. I think this can be dangerous but is on the right track. I would envision an anonymous auditing system where you get a random id assigned to your vote allowing for online validation. Although this probably would just provide a false sense of security as their are countless of ways to subvert this.

Yeah enough of my idealistic crap. We do it because it allows our politicians to give out fat contracts to loyal friends who will return that money in the form of campaign contributions. It the cycle of life for these guys. They need to raise absurd amounts of cash to stay in office and have devise many a creative way to do so.

Re:Any hope?

[rucs_hack]

what an intriguing mix of insight and paranoid nonsense.

Voting machines do not exist for people to buy elections. They exist because it monetizes the election process, allowing people to get wealthy by controlling a process that is required in a democracy.

The problem is that it's potentially so lucrative, that these guys are rushing into the process, talking up security, trustworthiness and stability, whilst simultaneously ignoring those same things in the interest of gaining the mighty buck, and the mightier government contracts. I don't doubt there are dodgy dealings being employed to gain those contracts, but election fixing? Be serious. You can't spend much money in jail, and they'd go away for a very long time.

Whatever, their aproach doesn't work, that's clear. They need to sort themselves out, or a new consortium of open source hardware and software bods need to step into this mess and offer decent replacements.

Re:Any hope?

[stiggle]

Why do you need instant reporting though?
The posts being voted on don't change for a few weeks after the elections, so its not as if you need an instant tally.

Votes should NOT be counted until the final polling stations have closed, otherwise the results from one station could affect the results in another. This could be a problem with the USA as it has multiple timezones, but they could just do their exit polls and then count the votes the following day and get the results in a reasonable civilised manner.

Re:Any hope?

[Smidge204]

I think "instant reports before the voting is completely over" should be an illegal practice. Exit polls are one thing, but publishing "official" vote results before everyone has finished voting? That doesn't strike me as productive to the democratic process.

=Smidge=

Re:Any hope?

[ArsenneLupin]

You can't spend much money in jail, and they'd go away for a very long time. They won't be doing that much time for a failed attempt at fixing the election. And, as has been shown in Florida, and 4 years later in Ohio, they will do no time for a successful attempt.

Think about it.

Re:Any hope?

[Splab]

Exit polls suck!!!!!!!! (+thousands of ! more)

We just went through an election, one of the national medias decided to make a prediction using exit polls and first 4000 counted votes (all voting places were closed). "If this holds this will turn out to be a major landslide". I had to find something else to do at that point :)
(No it didn't end up with any kind of slides, just the slightly above 50% on one side)

Re:Any hope?

[Splab]

Holy crap! How can anyone figure out whats supposed to happen with that butterfly ballot?

Here in Denmark you get a ballot thats a mile long and you put a mark at the line where you want to vote. (Multiple ballots when voting for more than one thing)

Re:Any hope?

[Cro Magnon]

You can't spend much money in jail, and they'd go away for a very long time.

To quote Aladdin in the Disney movie, "You're only in trouble if you get caught". Like most criminals, they don't expect to get caught.

Re:Any hope?

[Smidge204]

It is precicely BECAUSE exit polls suck that they are unlikely to influence the election. Also, if polls are closed throughout then that's not a problem either IMHO.

The problem is when you start announcing a "winner" before everyone has voted, which may disenfranchize people who would otherwise have voted for the "losing" candidate.
=Smidge=

Re:Any hope?

[TRS80NT]

Another factor which I didn't see in a quick scan of the other responses: Americans seem to need, even demand, instant gratification and feedback on an election. After a [too]long campaign, endless TV commercials, countless dinner-time phone calls and a wait in line we can't wait the day and a half it would take to hand-count a complicated ballot. I want to know if my guy won Now! Dammit. We see machines as a way to provide this.

Re:Any hope?

[Tacticus.v1]

May i suggest http://www.ebenhewitt.com/images/black-marker.jpg

Re:Any hope?

[Anonymous Coward]

Votes should NOT be counted until the final polling stations have closed, otherwise the results from one station could affect the results in another. This could be a problem with the USA as it has multiple timezones

No, this is a problem because some people on the west coast are stupid. I don't see why we should restrict the first amendment rights of the media just because some people are too stupid to know they should still vote after hearing east coast results.

Re:Any hope?

[Zigmun_Barsac]

Whew! That means that the machines cost over $5000 each! That's a lot of money for a PC with no printer.

Re:Any hope?

[OldeTimeGeek]

Does the rest of the democratic world have ballots that can be up two 8.5x14 pages thick?

The November 2006 California Gubernatorial election had seven statewide offices and twelve ballot propositions (Californians can enact laws through ballot propositions). This was just the statewide offices - it doesn't count any local offices, initiatives or municipal boards. And this was a small one. The California Presidential primary election in February, 2008 has seven state-wide propositions, with thirty two more in circulation and thirty two more in the final stages of verification at the Attorney General's office. Most of these won't appear until the November general election (where there are more propositions because, as a rule, more people show up to vote), but it gives you an idea of the number of issues people are asked to vote on.

And that doesn't even cover ballots for the sight-impaired, ballots in multiple languages, provisional voting which doesn't get counted until the voter has been verified and mail-in ballots. This is why we use electronic balloting.

Re:Any hope?

[simontek2]

Better question, why does the devices need a hard drive? It should be simple enough to run on a 4mb memory card.

Re:Any hope?

[Firethorn]

But that's still a solution looking for a problem.

Paper Ballots != punchcard ballots.

While supporters of paper ballots will probably argue as to the exact details (Pen! Pencil! Marker!), the fact is it's trivial today to print out scannable ballots, anybody who's attended school past elementory should be familiar with them, recounting is easy in multiple fashions, and they're far more cost effective - professionally printed paper ballots should cost less than ten cents each. You'd have to feed five thousand people through a single voting machine to make it worth it. To give you a hint: Assume a machine processes a votor every 5 minutes - that's 12 an hour, or a mere 240 people in a 20 hour period - far longer than polls are open. Even if you only give each voter a minute - that's still only 1200 people in an election.

Scalability? As long as you have ballots available, all you need is more pens, tables and maybe chairs.

Re:Any hope?

[Firethorn]

The most obvious advantage is the privacy it offers visually impared people who would otherwise have to depend on others to follow they wishes. (Kinda uniformed about this so don't kill me if I'm wrong and they have simple usable brail forms)

Some areas do have braille form. And I think it's hard to imagine a worse system for visually impaired voting than touch screen. At least with a keyboard you can have those little ledges to locate your fingers. You'd need a voice system - but then you either need earbuds or set the volume up high enough that everybody hears how gramps is voting.

A post above mentioned doing away with the secrecy aspect in the hope that they can audit their vote. I think this can be dangerous but is on the right track. I would envision an anonymous auditing system where you get a random id assigned to your vote allowing for online validation.

This is your employer* KrizDog, what's your id number? There's very good reasons no verifiability exists for voting.

Although this probably would just provide a false sense of security as their are countless of ways to subvert this.

Pretty much, if I control the backend I can make all the votes show up right while still subverting the numbers unless everybody submitted their votes to an independent counting agency, and sampling wouldn't be enough as long as the guy isn't trying to change a 90-10 vote into a 10-90 vote. Take a 48-52 for example. All you have to do is change that 48% into a 51%, and the 52% into a 49%. You changed 3% of the votes - not enough to detect for certain.

That was one of the objections to the diebold machines actually. It's system was to copy the vote table into another table - used the original table for showing individual voting district results, while the copied table was used for overall results. So as an auditor, unless I called up every single county and did the math myself, I wouldn't necessarily catch that somebody changed the copied table. Spot checks would be generally ineffective, as the machine reports the correct district results, while lying about the overall.

Right now that would probably be caught - but it would require somebody going the extra mile.

*Family, landlord, union boss, mob can all be substituted.

Re:Any hope?

[sumdumass]

West coast? This can be an east coast issue too. That's exactly what happened in 2000 in Florida.

You see, there is a difference from "I'll vote because I want someone to win even though I am extremely busy and it is inconvenient" and, oh, he has no chance of winning (or losing) so why bother stressing my life to vote for a loser (or someone who already won).

Of course this ties into the entire idea that they don't care about any other measure or candidates on the ballot. And with this seeming to be the popular theme on some elections, the stupidity levels exist not only on those who stay home but some of those that show up. Maybe forcing every voter who didn't vote the previous (unexciting election) year to vote provisional and then running a count between informed voters and people cheering a team on would help distinguish things a bit better?

Re:Any hope?

[roystgnr]

Voting is a simple enough proccess, why would anybody need a machine to do it?

Voting is an irrational process from a selfish perspective - the sum of the probabilities of me changing an election times the magnitude of the effect each election has on my own life is negligible even compared to the inconvenience of waiting in line to vote.

So when people do vote anyway, it's for some combination of:
a. irrationality,
b. zealotry,
c. entertainment,
d. altruism

People in category d are too rare to decide elections (although most of us in category b would mistakenly self-categorize there). Given the remaining three choices, our best hope may be to have elections settled by category c. And although following the debates and picking a good candidate is fun on it's own, most people in c want to see a fun minute-by-minute "horse race" during their election party at the end of the process. These people want instant official returns, and electronic voting seems to be the surest way to count a hundred million complex ballots with dozens of candidates and issues in a couple of hours.

What's worse, even the folks in category d are easily seduced by electronic voting - after the 2000 elections everybody in the USA understands the risks of ambiguous pencil marks and "hanging chads", and people who don't understand the risks of invisible databases and corruptible software think that electronic voting will fix all that.

Re:Any hope?

[bjorniac]

No. We elect a government and all the other ballots can go some other time. Why not have a presidential ballot that you do first, on paper, and then a machine in the next room for all the less important (ie local ordinances etc) stuff?

Re:Any hope?

[Anonymous Coward]

You can't spend much money in jail, and they'd go away for a very long time.

Several members of Diebold's board have been found guilty of fraud so clearly fear of jail hasn't stopped them before. There's also the presidential pardon card: do you think that they would serve "Scooter Libby" time or more like "Caspar Weinberger" time (if the whole thing wasn't declared a "state secret" first)? Finally, the amount of money involved is insane. The sheer magnitude makes the risk miniscule and the rewards epic: would you add a "bug" to your code for $500,000,000 (this problem applies to ANY e-voting system)?

The most important piece of evidence is the code itself. On the data collection side, a voting machine is trivial: at the end of the day you are counting button presses. Consequently, you have to be a bit suspicious of any solution that requires three "identical" sets of books (one contains the incoming votes and is used for producing pre-election verification reports, another contains the outgoing votes and is used for producing the election day results and the third is the mystery table that doesn't produce any reports at all). One should also be suspicious of a solution where all the built in auditing has been (at significant cost) disabled and which ships with tools for modifying records.

Ask yourself why a company that is exclusively interested in the profit from selling these devices would multiply their development costs by at least 10 to develop "worse" software?

If you are still a fan of e-voting at all, ask yourself how you knew that the machine that you voted on last election was not rigged? How could you ever be sure?

Re:Any hope?

[rucs_hack]

ask yourself how you knew that the machine that you voted on last election was not rigged?

I would go for incompetently managed over rigged, personally.

Re:Any hope?

[Opportunist]

It's not like the vendor of the voting machines would want to rig an election. What for? They'd get paid by either party anyway, so why bother?

But it makes verification harder. Anyone can take a stack of paper and recount. Anyone. You, me, that hick over there, as long as you can count and discriminate between the crosses (or holes, whatever the method).

Even if they can't count, unless it's by a very tight margin, just measure the stack.

Verifying the result of a voting machine without a paper trail is near impossible. You have to trust the vendor and the organisation using them.

It's not so much that it can be manipulated. What's the danger is that people will think it is manipulated, and you can't easily debunk them. When you have a paper-only voting process, simply dump the votes on the protester and tell him to recount.

I see the danger more in a loss of faith in democracy rather than actual manipulation. Whoever loses an election (or rather, his supporters) will cry foul and nobody will be able to show them (or, more important, the rest of the people who might listen) they're full of it.

Re:Any hope?

[DesScorp]

"They won't be doing that much time for a failed attempt at fixing the election. And, as has been shown in Florida, and 4 years later in Ohio, they will do no time for a successful attempt."

"They"? Who are "they" and what election did they steal? Proof? Something from a reliable source, please? And invoking Alex Jones is like invoking Hitler in usenet; it automatically disqualifies you.

Do you honestly think for a single second that most American Democrats would just lay down if they honestly, truly believe the election was stolen? Do you honestly believe the GOP got away with massive vote fraud in two elections, and the Democratic Party just sat on their hands?

Are you saying those Miami Herald investigations that declared Bush would have won the 2000 recounts was nothing but GOP propoganda? That they lied?

George Bush hasn't even been able to get the nominees he wants for many of his appointments. Are you telling me that those same Democrats that stonewall him at every turn just laid down for him in 2000 and 2004?

Come back to reality.

Re:Any hope?

[Anonymous Coward]

ask yourself how you knew that the machine that you voted on last election was not rigged?

I would go for incompetently managed over rigged, personally

I would take a little from column A and a a little from column B but that is beside the point that electronic voting cannot, almost by definition, be completely audited by anyone much less the average voter.

In a democractic election, anyone off the street is free to monitor the entire process: they can inspect the ballot boxes before and after, they can watch the ballot box, count the voters, observe opening of the box, observe the count, observe that the number of ballots counted matches the number of ballots cast, observe that number submitted matches the actual count, observe that the number submitted is included in the final tally and observe that the final tallies add up.

None-I repeat-none of these checks are available to electronic voters. In a black box system, you could flush your ballot down the toilet and have the same level of public scrutiny but you are supposed to rest easy knowing that the board of electors will verify that, at some point before and possibly again after the election, the toilet and the water in it met some sort of requirements.

Re:Any hope?

[doom]

Exit polls suck!!!!!!!! (+thousands of ! more)

In the 2004 election, many places were using electronic voting machines without any papertrail. The one and only way you could possibly detect any election fraud with those machines is by looking at descrepancies with the exit polls, and we did indeed see some strange descrepancies. Consequently, we've suddenly seen lots of people talking about how exit polls just suck.

Strange, that.

Re:Any hope?

[vbwilliams]

Actually, Diebold is the one with the "MS Access" solution...not ES&S. And this isn't about bullet-proof machines. It's about a Secretary of State making a splash based on the letter of the law, not the spirit of the law. What no one in the Slashdot crowd has every understood (nor does anyone else for that matter) is the economics of building a "bullet-proof" machine. Very simple question: If people are supposed to make and service these machines, what is an acceptable cost to building them, assuming whoever builds them still makes a profit to stay in business? In all these debates, that has NEVER come up. And it's funny...because it's the only driver in this whole issue. It's as simple as that...money. If jurisdictions really want a bullet-proof machine, they'd pay for one. But laws aren't written to allow people to vote. They're written to discourage people from voting. Knowing that, no one wants a "bullet-proof" voting solution. Knowing that, an affordable "bullet-proof" voting system is a paradox.

Re:Any hope?

[TapeCutter]

"counting OFTEN up to human judgement"

In a traditional manual system counting is ALWAYS adversarial, all sides count and watch - the winner is decided by CONSENSUS, if that can not be reached then the lawyers get their day in court.

Trivia: Thomas Edison created the first vote counting machine and tried to sell it to the US senate to count up/down votes. Most senators were quite justifiably offended by and suspisous of his machines.

Nothing has changed, a machine that counts your ballot cannot be trusted without implicitly trusting the people who supply and operate the machine. Manual counting does not have that problem, since trust in the elction system is critical to democracy - vote counting machines are far, far, worse than useless.

Re:Any hope?

[deniable]

Be careful with the words 'no one.' I've seen what the guys who got EFT machines into the European system had to do. It included serious security measures and Motorola providing chip schematics to the bankers. If an EFT machine can be built that tough then so can a voting machine.

The letter of the contract and the spirit it was written under are: if you change your machine, you'll let us know and decide if we have to certify it again. Changing enough around to justify a new model number should be an indicator that this needed to happen. What you don't understand is that companies live and die on contract language. I've worked on contracts that had damages of $200,000 a day for late delivery. These guys have no excuses.

And to your other point, I have a federal election in two days. I have to vote just like everybody else. I get a white piece of paper for the Senate and a green piece of paper for the House. That's it. We don't need machines to handle the voting. This race is close, we may not know the winner until Sunday.

Re:Any hope?

[vbwilliams]

I understand it completely actually. I used to work for ES&S. That $9 million is nothing compared to probably what they were losing by holding onto inventory that wasn't being utilized. And in the end, they won't pay a dime. They'll end up *leasing* (because NO ONE buys voting machines) the machines for some discount, then change order or service call the inventory up the wazoo to make up for the cost of the lawsuit. Always happens that way. Always. And anyone who thinks otherwise is living in a dream world.

Also, all the paper stock that ES&S uses is funneled through them. You don't get to use anyone else's stock with their machines. So guess what? Here comes the markup on the paper stock too. And the voting paper stock isn't cheap. And the AutoMarks are not touchscreens...they are basically inkjet printers that write directly onto the paper ballot. If there are arguments of inaccuracy, then the PAPER ballots can be hand-counted and examined with the human eye so no electronics can be of fault.

THIS is why I said the spirit of the law isn't being represented here. It doesn't matter if the machine is malfunctioning, has different code, etc. It doesn't matter. With ANY of the AutoMarks, the proof is ON THE PAPER. There isn't some magical electronic record. The receipt is the ballot. This is why this particular lawsuit is absolute garbage. In the end, it's the Secretary of State exerting muscle in a lawsuit that doesn't matter...it's not going to affect one way or the other the accuracy of vote-counting in any election...because for the most part, the machines have already been taken out of the equation. It's nothing but the state of California trying to be a hard-ass and the leader of the states when it comes to election *reform*...or whatever you want to call it. It's 100% politics...has nothing to do with preserving people's voting rights or anything.

This would actually matter if a touchscreen was the machine in question. AutoMarks aren't touchscreens...nowhere close.

At some point, people should hear the other side of the story. Believe it or not, a LOT of Secretary of States out there don't give a damn about the voter. I know this because in my previous position, I worked with at least 6 of them the last 6.5 years. They just give a damn about the voter's *perception* about how the system is working for them. This lawsuit does nothing for the voter except potentially take a relatively good voting solution out of their hands temporarily.

Re:Any hope?

[stiggle]

There is a difference between reporting the results eg. "I the returning officer for Camden County hereby state that in the election of 2007 the votes were cast as follows:....." and "I'm in Camden County with Fox News and it looks like Billy-Bob Walton has a lead in the counting of the polls"

One is the medias right to report, the other is the official results.
If the East Coast didn't release the results until the election is over, there wouldn't be a problem - which is oddly enough what almost every other democratic country does. The main problem in the US isn't the results, its the news channels exit polls. To the point where candidates have given speeches on winning or losing based purely on the numbers the media have given 'calling' the election.

Wait till the votes are counted properly and fairly. Your country depends on it.

Different Enough

[Tsuki_yomi]

If they are different enough for the company to give them a new model number, they are different enough to need recertification.

Re:Different Enough

[mrbluze]

If they are different enough for the company to give them a new model number, they are different enough to need recertification. And with widespread public suspicion on e-voting, the gov't will want to be seen to be doing the right thing.

Re:Different Enough

[leuk_he]

If they did not change the version number nobody might have noticed. Even not it it was sold with a buildin trojan. What does that say about voting with computers?

Even as an e-voting opponent, this seems harsh.

[bobdotorg]

I could understand Cal's concern if different IC's were used, or if code was re-flashed. But if the two machines had the same circuit diagram, same components, and code, this penalty seems zealous. I live in California, and it's painful to see bureaucratic zealots nominally on my side, but being far from reasonable. This particular error on the part of the voting machine company appears to be on the level of a failure to file necessary paperwork.

Re:Even as an e-voting opponent, this seems harsh.

[king-manic]

I could understand Cal's concern if different IC's were used, or if code was re-flashed. But if the two machines had the same circuit diagram, same components, and code, this penalty seems zealous. I live in California, and it's painful to see bureaucratic zealots nominally on my side, but being far from reasonable. This particular error on the part of the voting machine company appears to be on the level of a failure to file necessary paperwork. The contract didn't say "penalties only if re-flashed", instead it mentions any modifications needs to eb examined and approved. If you signed that contract you must be an idiot to do this substitution. You have to be strict or else you have more "diebolds". Any and all changes must be examined. All penalties assessed would be based on contract law. Paperwork is how a legalist society is run. It's not like jumping through hoops is new to government contractors.

Re:Even as an e-voting opponent, this seems harsh.

[deniable]

Do you want your vote counted by people who can't read a contract? We used to have client documentation requirements of two ring binders for some and three ring binders for others. If we did it wrong, we would have lost 10% of the payment for a 20 million dollar machine. You bet the requirements were checked and double checked.

This case also serves as a warning that California will not take any crap from the vendors. It may prevent any further 'mistakes.'

Re:Even as an e-voting opponent, this seems harsh.

[bobdotorg]

I could understand Cal's concern if different IC's were used, or if code was re-flashed. But if the two machines had the same circuit diagram, same components, and code, this penalty seems zealous. I live in California, and it's painful to see bureaucratic zealots nominally on my side, but being far from reasonable. This particular error on the part of the voting machine company appears to be on the level of a failure to file necessary paperwork.

The contract didn't say "penalties only if re-flashed", instead it mentions any modifications needs to eb examined and approved. If you signed that contract you must be an idiot to do this substitution. You have to be strict or else you have more "diebolds". Any and all changes must be examined. All penalties assessed would be based on contract law. Paperwork is how a legalist society is run. It's not like jumping through hoops is new to government contractors.

You play the 'D' card. And having an e-voting executive promise to deliver an election to a particular candidate offends me to no end. However the machines in question, machines for the deaf that print out a ballot to be scanned, do not offend me. Diebold's machines, as do most other e-voting systems do. Still, the DA seems overzealous.

A relevant quote from the article:
One condition of that certification, according to her office, was that no "substitution or modification of the voting systems shall be made with respect to any component of the voting systems ...

So it appears that what it might come down to is the definition of 'component'. Taken too loosely, a change in the Pantone color of the plastic of a button would clear the hurdle. Taken appropriately, well, 'appropriately' will be something for the courts to decide. It might all come down to something like a quote from a hero of mine: "It depends on what the definition of 'is' is."

Re:Even as an e-voting opponent, this seems harsh.

[seanadams.com]

Really? Where shall we draw the line then?

There is surely enough electronics in these machines that it would be trivial to conceal a circuit that changes its behavior depending on how various circuit boards are physically mounted in a chassis, even when all the connections appear visually equivalent.

Without visibility of the source code, we have no idea what it's doing under normal circumstances, much less when bits inside of it as physically rearranged. Hell, even with full schematics and source code, things could be easily hidden in production units. No matter what we do, we're taking their word for it.

Get rid of the machines.

Re:Even as an e-voting opponent, this seems harsh.

[jimboindeutchland]

I don't really think the point here is degree of the changes made on the product. The point is that California wanted to have certified voting machines. After all the conspiracy theories surrounding Diebold after the last election, I'm not surprised.

It's not good enough to allow this sort of thing to happen (especially when a Court overturn an election result) and people should demand clean and fair elections. Would you want to see the [insert hated political party here] win just because some guy couldn't be bothered filing some paperwork?

Re:Even as an e-voting opponent, this seems harsh.

[MLease]

This particular error on the part of the voting machine company appears to be on the level of a failure to file necessary paperwork.

Perhaps. However, when the issue is something as sensitive as voting machines, with all of the distrust of them we've seen and experienced (for good reason), it is imperative that everyone follow the letter of the contract and the law. As others have pointed out, with closed source software, it would be trivial to encode a backdoor or cheat function which would be enabled by what appear to be innocuous hardware changes, and the certification process would be completely sidestepped. Even if it is just a paperwork issue, and the result of error rather than malice, they need to get the message that absolutely no changes will be tolerated without the proper procedures being followed. I'm 100% behind CA's decision on this one.

-Mike

Re:Even as an e-voting opponent, this seems harsh.

[bobdotorg]

It's not good enough to allow this sort of thing to happen (especially when a Court overturn an election result) and people should demand clean and fair elections. Would you want to see the [insert hated political party here] win just because some guy couldn't be bothered filing some paperwork?

Goddammit. I can't believe I'm defending an e-voting company. I'm in absolute agreement that new and relevant revisions should be re-certified, but I believe that the penalties in this particular case are harsh. I would make the same argument if Dick Cheney was sentenced to six months in jail for jaywalking, spitting on the sidewalk, or flipping off a school crossing guard as he spit on the sidewalk while jaywalking.

All I can really hope for in this case is that the e-voting company proves their case that the machines are in every relevant way equivalent or better than the earlier revision. And in that case, if the machines are at least equivalent, then IMNSHO a just outcome would be for the e-voting company's liability be equal to all of California's legal costs in the case.

Yes, companies should be liable for fucking up. Intentionally or not. But if it's reasonable to believe that the FU was unintentional, and if in the end the FU was of no consequence, then I can;t see how any corporate entity should be liable for more than the state's costs of the investigation / court case.

Re:Even as an e-voting opponent, this seems harsh.

[jimboindeutchland]

Sure, I see your point there, the punishment is harsh.

I guess they want to make an example out of these guys.

Re:Even as an e-voting opponent, this seems harsh.

[Anonymous Coward]

The real concern is that it would have been so easy for them to comply, and yet they apparently ignored the requirement. So today, perhaps it is just a failure to file paperwork, but we can't allow that to grow into anything worse. If private companies want to participate in an integral part of our democratic process, they need to be prepared to be subjected to heavy, at times even unreasonable scrutiny. There has been so much corporate wriggling and funny business with voting machines in the past few years that I am not prepared to give voting machine companies the slightest bit of slack - this is the kind of business they chose to get into.

Re:Even as an e-voting opponent, this seems harsh.

[Anonymous Coward]

One must also keep in mind that this is just the initiation of legal action, and in no way represents the final decision that will be made. Perhaps the demand for damages is out of proportion to the offense, but on the other hand it is important for the state to take decisive action and show voting machine companies that they will not compromise when it comes to the democratic process. The important part will be whatever the court decides constitutes a "component" and a "modification". You correctly point out that these words are potentially fuzzy, and in the relatively new intersection of computers and the law, these are exactly the type of legal distinctions that must be made before we, as a society, can stand on firm legal ground when dealing with technology.

Re:Even as an e-voting opponent, this seems harsh.

[Antique Geekmeister]

I see your point, but this is one of the reasons "mil-spec" parts cost so much. You get what you ordered, not what some vendor decided you might like instead. And those small differences add up. Cable locatons affect air flow and thus cooling. A very minor component change, such as a minor network chipset change, can swap network ports. And failing to notify the customer of the change is a very, very bad practice, especially in a sensitive system, because when a technician opens it up and two otherwise identical systems with the same model number don't match, foul play is going to be suspected.

Re:Even as an e-voting opponent, this seems harsh.

[hidden]

I hate to tell you, but a change in the color of a key could be important. What if that change made the key unreadable to color-blind people? Or reduced the contrast so it was harder for a partially-blind person to read? These details can be important when you're dealing with something like a voting machine. Sure, maybe my scenarios don't seem likely, but they are possible, and thats why the government wants the chance to make the call, not the voting machine company. Notice that (at least the way I read this) the government of California isn't asking to re-certify trivial changes, they're merely asking to be notified of any change (no matter how minor) so that the government of California decides what is trivial, not the vendor.

Re:Even as an e-voting opponent, this seems harsh.

[Bartab]

Better idea: Get rid of the secrecy. As in secret ballots.

Once my vote is posted by my name, I can verify it. I can also vote from my laptop - or better yet, my cellphone.

So what if I can sell my vote? I've got no issue with that. Legalize and formalize the process.

Re:Even as an e-voting opponent, this seems harsh.

[mabhatter654]

government lives and dies by paperwork. Vendors know this, govt employees know this. After all, a speeding ticket is just not following the "paperwork". Or how about Taxes, filing those properly is just some "paperwork" as well.

When the military orders hammers they order an EXACT hammer, down to what color and finish... it's all very important to somebody so the specs have to be followed exactly, even if the hammer is functionally identical, that's not good enough. It's high time that computer and software people get the message and play by the same rules as other vendors when they produce their products. Software makers need to learn how to follow their OWN documentation and provide the exact documented service called for in the contract... when it comes to e-voting even 1 line of code errant is cheating, it would be as important as "just changing" a line item of your taxes because the form "works better" that way.

The FDA and IRS and Military and Casinos and Banks all demand EXACT procedures when lives or money is on the line.... I'd say VOTING is even MORE important that the rules be followed. It's a fundamental shift in how software is expected to be provided and operated that's been LONG overdue. The whole attitude in software that it works "good enough" so release and move on has absolutely no place in the e-voting market any more than in banking or running the space shuttle... a certain large software Making Software firm refuses to be bound by those kind of contracts even when it's the military doesn't mean their underlings can get away with it forever.

Re:Even as an e-voting opponent, this seems harsh.

[sumdumass]

Sure the company ultimately didn't follow the terms of the certification in the eyes of the California Secretary of state but was it because they "can't read a contract" or because the conditions in the certification was left open to interpretation? There is a legal interpretation of the certification that says "One condition of that certification, according to her office, was that no "substitution or modification of the voting systems shall be made with respect to any component of the voting systems" (from the article) and the question is what does that really mean. We haven't seen it, just a comment that is biased by a party to the suit. IT could be that the language actually allowed this if it was documented or something of the sort. It might have an acceptable modification clause after the part that says everything must be the same. It could also state that nothing can be changed in the context of the components, processors, and software itself and it leaves changed like this open.

From the quote, "with respect to the components" can appear as if the components are the same, the location doesn't matter or it can mean they have to be in the same place as when it was certified.

So this is something that will probably have to be played out in the courts and we would probably see some conflicting interpretations of the conditions in the time to come. I think we should keep an open mind on this until it plays out a little more.

Re:Even as an e-voting opponent, this seems harsh.

[uglyduckling]

But the story gives the rest of the clause:

...until the secretary of state has been notified in writing and has determined that the proposed change or modification does not impair the accuracy and efficiency of the voting systems sufficient to require a reexamination and approval...

So all they had to do was write to the secretary of state, explain that they had just moved a few things around, and wait for confirmation that the secretary agrees that the changes don't require reexamination. That's a pretty standard thing to have to do for any company that makes things for government or military. It doesn't matter if it's the component layout, colour, font on the front panel legends - if it's changed then you notify the changes. It's absolutely amateur and shows a total lack of understanding of the importance of the job they are doing if they don't know this.

I've done a bit of work on mil-spec hardware in the past and even as an engineer intern I was fully aware that every resistor and capacitor on the circuit board was made to a specification and couldn't be changed without notification. If this is how they treat hardware that can be easily inspected and compared with specs, I shudder to think what the underlying software is like.

Re:Even as an e-voting opponent, this seems harsh.

[aadvancedGIR]

It sounds a good thing to sell your vote, but you might soon notice that the price of your vote might simply be not losing your job at the first election, not being jailed at the second, and not being killed at the third (and you would have personnaly voted for each of these steps, so it will be perfectly legal).

Re:Even as an e-voting opponent, this seems harsh.

[mpe]

I could understand Cal's concern if different IC's were used, or if code was re-flashed. But if the two machines had the same circuit diagram, same components, and code,

In which case the onus is on the supplier to prove that there has been no change, regardless of the model number being changed. Even if the difference indicates some trivial change, e.g. colour of the casing, then it's up to the supplier to explain this and get the customer's approval for the subsitution.

This particular error on the part of the voting machine company appears to be on the level of a failure to file necessary paperwork.

If it's part of a contract to file necessary paperwork then failing to do so consitutes breach of that contract.

Re:Even as an e-voting opponent, this seems harsh.

[evilgrug]

Exactly. The changes they make aren't necessarily the issue, it's the fact that they made the changes in the first place and couldn't be bothered to read or follow the contract.

A good example might be the infamous Van Halen incident where the band did thousands of dollars to a venue after they were given brown M&Ms. The band's rider stated that M&Ms were to be provided but buried deep withing the technical paragraphs was a line stipulating that brown coloured M&Ms were to be removed. Obviously they didn't give a shit about the colour of their chocolate, it was just an easily-verifiable test to make sure the technical specifications of the contract were followed correctly and accurately. The actual breach of contract didn't matter.

Re:Even as an e-voting opponent, this seems harsh.

[mpe]

Do you want your vote counted by people who can't read a contract? We used to have client documentation requirements of two ring binders for some and three ring binders for others. If we did it wrong, we would have lost 10% of the payment for a 20 million dollar machine. You bet the requirements were checked and double checked.

How much could it have cost you if you had messed up, been taken to court and the judge had taken the position "If the defendent can't even get that simple thing right what else might they have got wrong"?

Re:Even as an e-voting opponent, this seems harsh.

[Anonymous Coward]

I could understand Cal's concern if different IC's were used, or if code was re-flashed. But if the two machines had the same circuit diagram, same components, and code, this penalty seems zealous.

Except that they changed stuff without letting the state know they did it and letting the state determine that "oh, you just moved some wires around." What if they had re-flashed the whole thing and never bothered to tell anyone?

Bigger question, though, is that they changed the model number. Why did the people buying them not say "wait, let me look at our list of approved models... nope, not here!" Did ES&S assert that the model had been approved (lol fraud), and the buyers just took them at their word, or did the buyers just not bother to ask?

Re:Even as an e-voting opponent, this seems harsh.

[mpe]

I've done a bit of work on mil-spec hardware in the past and even as an engineer intern I was fully aware that every resistor and capacitor on the circuit board was made to a specification and couldn't be changed without notification.

Presumably to allow a non engineer to check using a photograph.

Re:Even as an e-voting opponent, this seems harsh.

[betterunixthanunix]

The law is the law, and no modifications means no modifications. This is a voting machine, not some office computer -- any change might have consequences in terms of accuracy, or the ability of a third party to manipulate the vote. What if the rearrangement makes it easier for me to hack the machine, because it puts the motherboard in a position that is easier for my to connect my palmtop to? I'm sure that is a far-fetched situation, but we can't allow that kind of scenario to go unchecked when it comes to voting machines, especially with the track record that digital voting machines have.

Re:Even as an e-voting opponent, this seems harsh.

[goldspider]

This is California. With their budget situation, do you think they'd EVER pass up an opportunity to grab money?

Re:Even as an e-voting opponent, this seems harsh.

[DanMc]

But if the two machines had the same circuit diagram, same components, and code, this penalty seems zealous. I live in California, and it's painful to see bureaucratic zealots nominally on my side, but being far from reasonable. This particular error on the part of the voting machine company appears to be on the level of a failure to file necessary paperwork.

This might not be the only issue; just the only clear cut legal issue that the state can latch on to. If you're having trouble with a vendor, and you have a contract clause that's stated so clearly and then absolutely violated, you go to this point and kill them. We can't assume the whole story is here. FTA: California Secretary of State Debra Bowen says, "ES&S ignored the law over and over and over again, and it got caught." Hmm... This implies the model change may not be the only thing the state is upset about.

I'm upset about the cost of the devices. Doing the math on the reimbursement only, (not the penalties) CA paid nearly $5 million for 972 non-compliant devices = $5000+ per voting device. I bet the engineer who shaved $5 off the manufacturing process by changing the mounting brackets was really proud of himself.

Re:Even as an e-voting opponent, this seems harsh.

[necro81]

But if the two machines had the same circuit diagram, same components, and code, this penalty seems zealous.

ES&S thought the changes were substantial enough to give the resulting device a new name (A200 instead of A100). Did they give it a new price, too? Plus, they didn't go through the motions to have someone verify that the same code was, in fact, used. Despite their claims, without having an independent tester do that verification, who knows what they flashed on there.

Re:Even as an e-voting opponent, this seems harsh.

[TimTheFoolMan]

You're probably not aware of this, but one of the early prototypes of an accessible voting system told the user (via audio), "Press the Red button to select this candidate. Press the green button to hear the next candidate."

With that kind of history from some of the vendors (that was not an ES&S product), it's easy to see why the legislation would get nit-picky.

Tim

Re:Even as an e-voting opponent, this seems harsh.

[forkazoo]

I hate to tell you, but a change in the color of a key could be important. What if that change made the key unreadable to color-blind people? Or reduced the contrast so it was harder for a partially-blind person to read? These details can be important when you're dealing with something like a voting machine. Sure, maybe my scenarios don't seem likely, but they are possible, and thats why the government wants the chance to make the call, not the voting machine company. Notice that (at least the way I read this) the government of California isn't asking to re-certify trivial changes, they're merely asking to be notified of any change (no matter how minor) so that the government of California decides what is trivial, not the vendor.

It's not just that California is "asking." It's that Diebold agreed, in writing to those terms. Some people may feel California is asking something unreasonable. Fine, I disagree, but fine. Doesn't matter if California asked that all voting machines have pictures of sad clowns on them. Diebold agreed to California's terms. Then it ignored them.

Re:Even as an e-voting opponent, this seems harsh.

[Firethorn]

Presumably to allow a non engineer to check using a photograph.

More than that. Military hardware is frequently built to survive conditions not experienced by consumer machines. Something as 'simple' as moving a cable could be the difference between the part survive an EMP blast and still be operational and frying instantly.

Rather than assuming an operating temperature range of 40-100F, it might have to survive -30 to 190F.

Substituting a non-shielded cable for a shielded cable, even where civilian specs calls for unshielded, can result in betraying RF emissions allowing enemies to gain valuable information.

Re:Even as an e-voting opponent, this seems harsh.

[sumdumass]

So all they had to do was write to the secretary of state, explain that they had just moved a few things around, and wait for confirmation that the secretary agrees that the changes don't require reexamination. That's a pretty standard thing to have to do for any company that makes things for government or military. It doesn't matter if it's the component layout, colour, font on the front panel legends - if it's changed then you notify the changes. It's absolutely amateur and shows a total lack of understanding of the importance of the job they are doing if they don't know this.

I think your missing the point. Or maybe missing the forest for the trees.

You see, "all they would have to do", would depend directly on the definition of component, change, and the wording putting the two together. Because if the changes they made wasn't required to be reported and ok'd by the secretary of state, then they wouldn't have to notify them. If they defined components to mean the drive, IC's, processors and so on, the configuration of them might not be covered and therefore the changes they made might not have needed to be reported. So while it might be that any change at all might have needed to be reported and ok'd first, it is even more likely that one party interprets it in a way that they didn't need to notify anyone.

When this thing progresses, we will get more information about the specifics and even hear the defense of the company in question. Then we can make the claims of guilt or innocence. But until then, it is wild speculation based on a biased opinion not fact.

I've done a bit of work on mil-spec hardware in the past and even as an engineer intern I was fully aware that every resistor and capacitor on the circuit board was made to a specification and couldn't be changed without notification. If this is how they treat hardware that can be easily inspected and compared with specs, I shudder to think what the underlying software is like.

Nobody is claiming that they changed resisters or ICs or anything. They are claiming they changed the location of the same components for whatever reason. And another thing is that no one is claiming that this is Mil-Spec hardware. At least not in the definition you are familiar with and making it out to be. The requirements to keep things the same are an add on condition to the original certification of the unit. And this requirement relies heavily on the interpretations of what component means and what changes to components means according to the condition set on the certification. When we see that condition, or hear what it says from an unbiased party, we will know more about it.

I have to ask, are you attempting to make this look evil in the publics mind with the intent of biasing any jury that might sit over it because you have some alliance with the California secretary of state or dislike voting machines that much? Or are you drawing too much from your own experience that doesn't directly translate to this situation with the amount of information we have. Because you see, the former statement is a biased opinion pushed as fact in the form of a question. The later statement is is still a question but it doesn't push a bias or malicious intent on your part. They ultimately ask the same thing, "are you wrong" but one furthers and agenda while the other implies a mistake. What we know so far is a biased opinion from someone furthering their agenda. Let get all the facts (or at least a few more) before we become their cronies unwittingly.

Re:Even as an e-voting opponent, this seems harsh.

[roystgnr]

This case also serves as a warning that California will not take any crap from the vendors.

I'd say it serves as a warning that the only way California can detect vendors' crap is by noticing new model numbers and obvious visible physical changes to certified software. The only way they can even hope to detect invisible changes to certified software is with a chain of trust which has again been proven to be missing most of its links.

It may prevent any further 'mistakes.'

It may discourage any further mistakes, but I'd say it only encourages 'mistakes'.

Re:Even as an e-voting opponent, this seems harsh.

[Anonymous Coward]

/me coughs discreetly: As you say, sir, but perhaps you meant ES&S, not Diebold.

Re:Even as an e-voting opponent, this seems harsh.

[uglyduckling]

Woooahhh less of the conspiracy theories. This is a discussion site, I'm discussing - why are you so bothered? In any case I'm British and living in England so I have no desire to influence anything. I am interested in discussing because US elections have a worldwide impact, and I'm sure that the UK will go electronic eventually so what happens in the US again will affect me.

I didn't claim it's mil-spec. I used something called an analogy. I have experience developing mil-spec and drew on that experience. It's not a huge leap of the imagination to see that developing something for government elections might be similar to developing for military use in terms of the level of attention to detail and adherence to contract specifics. All I'm saying is that in my experience when you're developing hardware/software for something critically important, if you sign a contract saying that you will notify any changes, that means you will notify any changes. It's as simple as that. I'm not making any kind of 'wild speculation'. Either the story is false - they did in fact notify the changes - or else the basic details are true, in which case it's a major problem, whatever defence they may think they have.

Imagine (another analogy here) you asked me to build a house to your specifications, and to discuss any changes with you. Does that require interpretation? Maybe the first time I asked something absolutely trivial you might clarify things, but that would be your choice. It would be entirely wrong for me to forge ahead and decide for myself what might be excluded from the terms.

By the way, I'm not that familiar with US law, but I'm pretty sure this wouldn't be a trial by jury. I'm happy to be corrected.

Re:Even as an e-voting opponent, this seems harsh.

[sumdumass]

Woooahhh less of the conspiracy theories. This is a discussion site, I'm discussing - why are you so bothered? In any case I'm British and living in England so I have no desire to influence anything. I am interested in discussing because US elections have a worldwide impact, and I'm sure that the UK will go electronic eventually so what happens in the US again will affect me.

I didn't state a conspiracy theory or an agenda that your pushing. I simply illustrated how it gets that appearance when you rush to judgment based on someone else's biased opinion. Your discussion it entirely welcome but the appearance of ignoring details isn't. If I mistook your statement and it should read that "all they would have had to do was" instead of "all they did have to do", then My bad. The difference depends directly on the point I was making.

You see, you glossed over my point that the conditions in the certification could be interpreted entirely differently then what the California secretary of state claims and you went straight to the "So all they had to do was". The point I was making is that they probably didn't have to do even that, or their interpretations of the clause could be to that effect. If the conditions set forth in the certification actually say something different then what the government thought/thinks, they didn't have to inform anyone of the changes made and they didn't need to wait for any approval for any reason. This would be a big mistake based on the misunderstanding of what a component means and the limits to that definition set on them.

I didn't claim it's mil-spec. I used something called an analogy. I have experience developing mil-spec and drew on that experience. It's not a huge leap of the imagination to see that developing something for government elections might be similar to developing for military use in terms of the level of attention to detail and adherence to contract specifics. All I'm saying is that in my experience when you're developing hardware/software for something critically important, if you sign a contract saying that you will notify any changes, that means you will notify any changes. It's as simple as that. I'm not making any kind of 'wild speculation'. Either the story is false - they did in fact notify the changes - or else the basic details are true, in which case it's a major problem, whatever defence they may think they have.

Don't take that last part I said to heart. It was simply an illustration on how making a strong statement based on one sided biased information ends up giving the wrong impression. I hoped that I had explained that enough immediately following it. I also understand how you would draw from your own experiences just as I have in the past. But in doing so, you were creating a situation of interpreted facts that are really biased opinions. That is something I think you should think about a little because it is so easily done. Think about how this line of logic goes with another illustration, the cops arrest people who do bad things, someone got arrested so he must be a bad person. Therefore when I'm on a jury, how to I interpret the events leading up to the arrest of a bad person?

There is little difference between something like that and basing your opinion of incompetence and guilt based on the one sided view of a biased party. I could give more examples to, I have them with race, religion, or however you might want to see it. Basically, keep an open mind until you have heard both sides of the story and can close your mind in an informed way.

Imagine (another analogy here) you asked me to build a house to your specifications, and to discuss any changes with you. Does that require interpretation? Maybe the first time I asked something absolutely trivial you might clarify things, but that would be your choice. It would be entirely wrong for me to forge ahead and decide for myself what might be excluded from the terms.

Re:Even as an e-voting opponent, this seems harsh.

[forkazoo]

Egad! As you say, sir. I'd like to say something on the matter, but I'm busy at the moment trying to extract my foot from my mouth...

I guess it's just force of habit whenever there is a voting machines discussion.

Re:Even as an e-voting opponent, this seems harsh.

[mrcpu]

Even if you had the source to the program, you have no guarantees that the processor will use them correctly (as specified), or that the processor doesn't have microcode that basically ignores whatever you wrote. You'd also need to verify the entire compilation toolset as well...

One of two things happened

[Chris Snook]

a) They didn't think it was that big of a deal.

b) They forgot.

The actual error isn't terribly worrying, but the process failure that led to the breach of their contract, especially for something that could have been complied with quite easily, is not the sort of thing you want to see going on at a company that makes closed source voting machines.

Re:One of two things happened

[jimicus]

The actual error isn't terribly worrying, but the process failure that led to the breach of their contract, especially for something that could have been complied with quite easily, is not the sort of thing you want to see going on at a company that makes closed source voting machines.

It's not the sort of thing you want to see going on with ANY company which provides closed-source solutions - either software, hardware or a combination of the two.

Now what would happen if a medical equipment manufacturer tried to pull a stunt like that?

Re:One of two things happened

[simong]

If either is the case then they deserve every penny of the fine, and should be banned from providing hardware or software to any voting system anywhere. In their business there is no excuse for mistakes or oversight. If something is changed there should be a process of documenting that change, accepting the change and, most importantly in this case, validating that change and getting it certified by the controlling bodies. There should be a procedure for that, all written down and set into motion as soon as one bolt is loosened.
If voting isn't that important, let's give up now, and you can let Halliburton or Microsoft or Starbucks directly decide how your taxes are spent without the pretence of having elections.

Re:One of two things happened

[forkazoo]

The actual error isn't terribly worrying, but the process failure that led to the breach of their contract, especially for something that could have been complied with quite easily, is not the sort of thing you want to see going on at a company that makes closed source voting machines.

Indeed. Changing the position of a bracket and whatnot is a pretty minor design tweak. OTOH, this is a voting machine, and it seems so fundamentally obvious that every employee at every level should have been indoctrinated with the idea that no uncertified changes can ever be made. If people are making minor changes without a very careful process, then you have a culture which clearly wouldn't have considered it a big deal to "tweak" the software in the machines. Nobody would raise a red flag at it, and anybody with access could have modified the software however they like post-certification.

It seems like the sort of corporate culture specifically designed to deniably engineer an untrustable machine.

Checks and balances

[TheBearBear]

SOME LINES FROM THE ARTICLE...

Does relocating two circuit boards, rerouting several internal cables and changing some mounting bracket supports mean an e-voting device must be recertified to meet state e-voting requirements?

The company also contended that the changes to the AutoMark A100 were so minor that ES&S was not required to submit them for review.

The only changes made to the devices were minor engineering modifications, according to ES&S.


Let me answer the question at the beginning of the article with a resounding YES!!!!!!!!!! YES YES YES! What if the software was written to act differently (cheat) if a bolt was in a certain place, if the color of some paint was different, or if something else was a cetain height? A company can just say "these are just minor changes that has nothing to do with the operation". You see, the contract was written to cover things like this. I am not saying the company had ill-intentions, but if they did violate the contract it's just stupid and - i guess if I can stretch it - a bit suspicious.

Re:Checks and balances

[deniable]

Here's a fun one. If they're minor changes, why did they conspicuously change the model number? I bet that's the only reason they got caught. The contract says model 100 and no changes, but we'll give them the 200 anyway. Some official looked at the logo on the box and thought, "that's not what we ordered." If they'd called it a 100A in the small print and left the 100 on the front, they would have avoided this mess.

Re:Checks and balances

[Kierthos]

And they still would have been in breach of contract. It doesn't matter what their intentions were... a better machine, more easily serviceable... doesn't matter. They broke the contract, they pay the penalty.

Re:Checks and balances

[deniable]

Absolutely, but they then hung a big 'catch me' sign on it. Breaching contract is one thing, but being stupidly obvious about it is another.

Re:Checks and balances

[Kierthos]

Well, legally, it would actually be worse if the had made the modifications, tried to cover it up, and then still gotten caught later.

Re:Checks and balances

[iamacat]

However, illegally, they would get away with it.

Re:Checks and balances

[remmelt]

Doesn't that make you all warm and fuzzy inside when thinking of electronic voting? That we just established that they can easily get away with it?

Re:Checks and balances

[bflynn]

Not necessarily. Part of the question lies in what constitutes the "voting system". ESS obviously considered the voting system to lie in the functional operation. Somebody in the state government considered it otherwise (or saw a chance to make some hay).

Regardless, the courts will sort it out and say who is wrong and right. An interesting estoppel argument for ESS is that this machine was clearly labled differently. If they explained the change and the counties bought the machines anyway, the state could be stopped from claiming any damages since the voting officials in the counties didn't believe it was a change to the voting system either.

Re:Checks and balances

[noidentity]

The company also contended that the changes to the AutoMark A100 were so minor that ES&S was not required to submit them for review.

If they were minor enough to not be worth re-certification, the changes shouldn't have been made. When you're dealing with software that must be secure or is used in life-critical situations, you simply don't make minor modifications in the first place. Unless you're a company selling voting machines to the US.

It may be more serious than obvious

[Anonymous Coward]

Sometimes simple modifications substantially weaken security. The relocated circuit boards could make it easier to swap chips, or make targeted DoS attacks which can easily alter elections easier to effect. It's well know that most election districts have a history of voting for candidates of a certain party. If you knock a bunch of machines offline in just a few of the ones for the opponent, you can cause the lines to be long enough fewer people will vote, and unless it would be a landslide, the election results change. As for how to knock the machines offline? Instead of needing a NERF gun, perhaps the changes allow something as simple as a high power white noise generator with an antenna beneath a person's clothes to do the trick.

Re:It may be more serious than obvious

[iamacat]

You are extremely naive if you think the state did this level of testing. Most probably the boss just cast 5 ballots, verified the results manually and declared that the machine works correctly. Even if they did honest testing at the usual government's level of sophistication, it was not anything that would be changed by rearranging boards.

Re:It may be more serious than obvious

[iluvcapra]

California's testing and qualification of voting machines -- this time around, at least -- was quite formidable, including tiger teams to hack the machines. The report from the testers was the reason they all got sent back in the first place.

Re:It may be more serious than obvious

[jimmydevice]

If the state isn't doing the tests spelled out in the contract they are just as culpable. Since they caught this re-configuration of the hardware, they have obviously examined the hardware for modifications. I suggest looking at the law as written to understand their decision and the test procedures.

Re:It may be more serious than obvious

[KDR_11k]

That and rerouting cables could change signal latencies which could change the system's behaviour, especially if the system is designed to be vulnerable to such changes.

No Change To The Voting System...

[LEX LETHAL]

They didn't feel the need to notify in writing because there was no change to the voting system. There was, however, a change to the internal hardware layout.

Re:No Change To The Voting System...

[Anonymous Coward]

So the hardware is not a part of the system?

Re:No Change To The Voting System...

[EmagGeek]

THANK YOU. Changing the physical location of a system board within the chassis doesn't do anything to change the system. FFS, people need to get a freakin' grip.

Re:No Change To The Voting System...

[dave420]

It's up to California to figure out if anything important changed, not ES&S. Democracy is kind of important?

Further proof...

[Matti-han]

...that to be chosen to govern something does not mean actually having to know anything about it. At least from what I read it seems all they did was re-arrange how the components were mounted inside the machine. Would they sue if the company had left everything the same but used different sized screws? I'm glad someone is paying attention, but come on, do a bit of homework. If there is no actual functional changes, I could see why the company might not have thought of notifying the state.

Even given they should have notified them, is this really how to resolve this? Please, go on using my tax dollars to fatten more lawyers. It's not as if you couldn't just RETURN the ones that are non-certified and taken your business elsewhere like every other rational consumer.

Whatever the hell is going on, I smell 'idiot' all over this. I don't know or even care to guess who or on what side, but this just reeks of it. I'm not for or against electronic voting machines per se, I'm against people with the technological understanding of a retarded hamster being in charge of them.

Re:Further proof...

[statemachine]

It's not as if you couldn't just RETURN the ones that are non-certified and taken your business elsewhere like every other rational consumer.

It looks like that's exactly what California is trying to do. Plus, it also costs the counties and state money on top of the original price to replace these machines. It seems reasonable.

Re:Further proof...

[pegdhcp]

That is also proof that, to be able to talk about something does not mean you know shit about it.

What if they have some solid state gyros, that act differently based on said boards' positions. Or if that is so complex to accomplish for some idiots, who cannot fill in some papers; What if the PCB is designed in such a way that, when you use a one size bigger than standard nut, it acts as a jumper and changes results...

Re:Further proof...

[KDR_11k]

They had a contract. The supplier failed to follow it. Breach of contract is handled like this, I doubt the contract had anything about "just return it and go elsewhere" in it (also I'm not sure govt spending can be redirected as easily as that, remember that there's the whole lowest bidder thing to go through). Even if, that would cause additional costs for the govt because of the additional work to get another machine made. If a store sells you a product that's not what you told them you want that's fraud, if a supplier sells you something that's not what the contract specified that's breach of contract.

Amazing, how can you be this stupid

[SmallFurryCreature]

This is EXACTLY what happened with all those chinese product safety scandals. A safe 'certified' product gets produced in China, someone there decides to change something, and BAM the product turns out to be unsafe.

Certification is meant to be "I seen this product, I tested it, it is safe". If you then CHANGE that product, that means the test is no longer valid.

And yes, that is down to the size of the screws. In this case that would matter a great deal, voting machines are supposed to be tamper proof. Change the screws and it might be a lot easier to open all of a sudden.

If you work with products that are certified, then you must keep the product the same. Those are the rules, it is in the contract.

Really, with the recent stories from China I would think nobody would be stupid enough to think it a good idea when products are changed on the production line.

It don't matter that the changes may not have an impact, HOW ARE WE SUPPOSED TO KNOW.

The deal with this kind of situations is, you produce a product in X form. That is form is tested and gets certified. If you then change it, it has to be retested and recertified because without it that product has suddenly become untested and your word isn't good enough or we would have gone through the first testing and certfication in the first place.

Do you trust voting machine companies? You must be a diebold stockholder.

The finer details

[JonnyO]

If only they were this stringent about the code these machines are actually running. I guess it's just a case of see no evil...

Re:The finer details

[myspace-cn]

THEY being the EAC? Time to do some journalistic research on that. (not knocking you, just springing off ya. Don't guess anymore, use intuition as fact. YOUR LIT!)

Not modified but upgraded

[edwardpickman]

The new Republican only software service pack was an upgrade to improve machine performance and simplify the voting process.

Re:Not modified but upgraded

[RuBLed]

aha, so they added a cash dispenser, makes sense now...

(*deep breath*... haaayyyy....)

Re:Not modified but upgraded

[myspace-cn]

Bud. Your lit. Hella lit. The TRUTH.

Re:Not modified but upgraded

[pdurod]

Forgot the "special interests" service pack 5.0

Screw up.

[TheBearBear]

For some of you guys who are saying "OMG it's just a little screw what are they fussing about?!", let me give you an analogy. Let's say you're gonna have a secret meeting and you and your cronies have everything planned down, to the last detail. Every little thing, even the color of their shirts that they're supposed to pick up at a certain place at a certain time. The time, the day, what to bring, everything. Now suppose your guy shows up a second late, or wears a different color shirt. He says "Aww man i couldn't get it so i just dropped by walmart right quick". What do you do at this point? Drop everything and leave? Or let the color of the shirt bother you?

What if I were to say that your buddy couldn't get his shirt on time because he was pulled over by the FBI at the last minute? This is why you made those crazy rules in the first place, so that you'll be able to catch anything out of line. That's insurance!

Re:Screw up.

[RuBLed]

I did not get it.. could we try it again.. this time using cars...

Why Computing.org and not BRADBLOG?

[myspace-cn]

Computing dot whatever ain't been kicking ass and taking names and telling like it fucking is like BRADBLOG has! http://bradblog.com/ DONATE We been saying all along the answer is paper ballots hand counted with public oversight.

What the hell

[MrCopilot]

I've been seeing overly paranoid comments on SlashDot for years. What the hell is this its no big Deal crap.

It is this simple.

The law REQUIRES Cerification of the Voting Machine to be used/sold. ESS had the A100 certified. They are allowed to sell the A100 in CA

ESS made a newer model the A200 and sold them uncertified to districts.

I don't care what the changes were, You put a sticker on it that wasn't there during certification its uncertified. PERIOD. Finish engineering the damn thing before submitting it for certification.

Let this be a lesson to the makers of these types of machines. ONLY CERTIFIED VOTING MACHINES are legal.

Frankly, I'm disappointed with you guys for your wishy washy interpretation just because we are a bunch of engineers doesn't mean we don't have to take their side when they violate the law. Especially wjen it comes to something so vitally important to our democracy.

I couldn't agree more with this comment from the Sec of State.
"ES&S ignored the law over and over and over again, and it got caught," Bowen said in a statement. "California law is very clear on this issue. I am not going to stand on the sidelines and watch a voting system vendor come into this state, ignore the laws and make millions of dollars from California's taxpayers in the process."

Thank You, that is all.

Re:What the hell

[dabadab]

Let this be a lesson to the makers of these types of machines

No, let this be a lesson for the voters: if something as obvious as a modell with physical alterations and - for crying out loud! - a different type designation could be sold to the districts bypassing all the security measres that should have been in place preventing this, then how do you trust these exact same security procedures to catch subtle modifications of the software?

Re:What the hell

[tcampb01]

Did I miss something when I read the story? I'm failing to see how this is ES&S' fault.

Let me see if I got my facts straight... (a) California certifies a model A100. (b) ES&S introduces a "new" model called the A200. (c) The A200 is NOT a certified machine. (d) The voting districts approach a vendor and ask to buy a product (A200 e-voting machine) from a vendor. (e) The vendor sells them what they asked for.

Assuming I didn't miss any details, the fact that the districts asked to buy e-voting machines which were not certified as an 'approved' model is hardly the vendor's fault.

This is NOT a problem with ES&S or the A200 or even the old A100. This is a problem with buyers of e-voting machines in California. If anyone needs to be scrutinized and punished here, it's them. Even if ES&S had (accurately) told the buyers the the A200 was essentially just an A100 because it uses the same boards and wiring but the boards are mounted in a different location, they still didn't represent it as an A100 or as a 'certified' machine. The decision to assume it was ok to buy a different model was up to the buyers. Nobody forced this on them.

Had ES&S made changes (no matter how trivial) to the certified A100 and then CONTINUED to sell these modified versions still as model A100, then that would have been a misrepresentation.

Re:What the hell

[Anonymous Coward]

"...just because we are a bunch of engineers doesn't mean we don't have to take their side when they violate the law." I'm assuming you meant: "...doesn't mean we have to take their side..."

But it isn't the engineers that did this. Sure, they design the thing, but it's the managers who decide everything else. i.e. how much time and resources (or lack thereof) the engineers have, what to design, which model to manufacture, quality standards, where to sell it, etc. Managers even make engineering decisions under the guise of accounting, or whatever the dilbertesque excuse of the day is.

Re:What the hell

[MrCopilot]

Let me see if I got my facts straight... (a) California certifies a model A100. (b) ES&S introduces a "new" model called the A200. (c) The A200 is NOT a certified machine. (d) The voting districts approach a vendor and ask to buy a product (A200 e-voting machine) from a vendor. (e) The vendor sells them what they asked for.

(d)The districts approached the vendor to purchase a certified machine. ESS is listed as a vendor with certified equipment. District donot have the responsibility to know the model number. They ask for a California certified voting machine. They were sold uncertified machines. Fault ESS.

Where do you get the idea that districts asked for the A200?

(e)As I pointed out above, no they didn't, they bait and switched the districts. Later saying they thought it was no big deal and didn't "think" they needed to re-certify.

Re:What the hell

[MrCopilot]

Exactly, sorry about the jumbling.

What it shows is the utter stupidity of the management at ES&S that made them think they could alter the machine because the Engineering team said "In order for production of this to go forward we need to move a few things around" and management said well, we'll send it for federal testing, but not State. Thus breaking the Law.

This machine has features for the disabled/blind, Anybody know if the Audio feedback is through headphones or a speaker? Just wondering.

Re:What the hell

[Anonymous Coward]

This reminds me... I haven't received a letter from the office of the Secretary of State concerning California voting. I wanted to find out if they were going to have non-computerized voting available. What they had the last time was abysmally bad. How much money has been wasted on voting machines since 2000? And it seems to be for a system that doesn't work very well at all.
      If they keep pursuing the electronic voting machines, we're all going to be voting absentee in short order.

Re:What the hell

[Anonymous Coward]

Physical modifications may be obvious to you, but they're not obvious to me. I've never seen an electronic voting machine. Not even a picture. The election officials may not have ever seen one, either. And, even if they had been seen, how do the election officials know whether the altered model has been certified? If it hasn't, what is the official supposed to do - are they set up to handle paper ballots, because the voting machine is uncertified? Or should they turn away voters?

If they go to paper ballots, then aren't they wasting a bunch of money that was supposed to be saved by going electronic? (or do the electronic machines print out ballots, meaning that they're just a different UI for the same old paper ballot?)

Retarded Hamsters

[Harmonious Botch]

Parent's last paragraph is the most coherent comment on this article so far. Someone please mod him up.

Re:Screw up...with cars

[TheBearBear]

Ok, let's say the car I wanted from you met with specifications and is certified. Everything was specified, down to the type of screw (let's call them XY700 screws), type of paint, color of paint... Now let's say you decided to get your screws from a different vendor because it's cheaper for you, instead of getting ones from the vendors we specified. BIG DEAL, right? OH YEAH, BIG DEAL. That's because the XY700 screws from those other vendors do not meet metal fatigue resistance our standards. So let's say one of our engine/steering components come loose during driving because the screw broke, causing the car to flip or something. And it turns out that was because the screws use were of lower grade (same metal, but manufactured with a different process). BIG DEAL ALRIGHT!!!!! Those are things engineers look out for. It may seem trivial to the laymen, but alot of money is at stake. Can't substitute sugar for corn syrup at this level.

States that bought them may be also at fault

[wvmarle]

Not knowing the details, but:

- there were two types
- one was certified
- the other was sold

So the states bought knowingly an uncertified machine?

Sue them all!

[Anonymous Coward]

Why didn't they handle this in a more constructive and positive manner? To us Europeans this sounds all to much like a sue-em-while-you-can thing, typical of what's wrong with America these days.

Look at their track record.

[Qem]

http://www.news4jax.com/politics/3890292/detail.html "The software is not geared to count more than 32,000 votes in a precinct. So what happens when it gets to 32,000 is the software starts counting backward," said Broward County Mayor Ilene Lieberman. The article says that they'd known about the problem for two years and failed to fix it. http://abcnews.go.com/US/comments?type=story&id=2646802 Randy Wooten figured he'd get at least one vote in his bid for mayor of this town of 80 people even if it was just his own. He didn't. Now he has to decide whether to file a formal protest. http://backslash.slashdot.org/article.pl?sid=06/08/01/191235 The Open Voting Foundation's disclosure that only one switch need be flipped to allow the machine to boot from an unverified external flash drive instead of the built-in, verified EEPROM There has been tons of mishaps with those machines reported on slashdot alone... I certainly don't blame them for throwing the book at them and fining them for all their worth. It certainly sends across the message that the voting system is not to be fucked with and hopefully it can help prevent situations like the above.

Re:Look at their track record.

[Fujisawa Sensei]

Considering their track record, why are they even buying additional machines from them to begin with?

Small changes?

[Thanshin]

This is computer science.

We use 1 and 0 as the smaller element to store all information.
We use 1 and 0 as true and false.

There isn't such thing as a small change.

I know what I would do...

[Belial6]

If I was a voting machine vendor, and I wanted to hide a hack, er... miss an accidental bug, in the original hardware, I would just have a ground point that enabled the alternate code. Then the only modification needed would be to leave off an insulating washer. Far less than the modifications done to this machine.

It IS a big deal

[TheLink]

The company obviously doesn't treat elections very seriously and should be smacked down.

An election doesn't just have to work right, it has to be SEEN to be working right - that is PART of the "deliverables". Otherwise people may rightfully get pissed off.

Say in an "old school voting system" you had a company in charge of transporting ballot boxes from the booths to the counting stations, and one of the trucks took a "minor detour" on the way, maybe for the convenience of the company or the employees (take a leak or buy a drink etc).

Sure, nobody might have tampered with the stuff, but the elections get "damaged".

How damaged who knows. The eventual losers could kick up a big fuss. You might piss off millions of voters.

The company obviously doesn't deserve to be in the election system business.

The USA spends so much money in Iraq on "regime change" AKA picking the leaders there, but when it comes to picking the leaders back home - "it's only a minor modification" or we'll let Diebold's rejects who wouldn't be able to make ATM machines build voting machines for us.

Re:It IS a big deal

[freedom_india]

Nope. It would NOT be smacked down like in WWF.
It would be "settled".
The company would pay a visible compensation and an invisible large contribution to campaign funds and the matter would be closed; without the company ever admitting it did wrong.

During 1800s many local counties had laws which prevented convicted companies from doing/establishing business in that locality.
But then now convicted companies like Deibold http://www.heraldtribune.com/apps/pbcs.dll/article?AID=/20031217/APN/312170634 and Microsoft http://www.vnunet.com/vnunet/news/2118681/microsoft-convicted-software-piracy
  not to mention Bechtel http://www.globalpolicy.org/security/issues/iraq/reconstruct/2006/0905profiteers.htm Custer Battles do business with impunity in all states.

When am convicted of a crime, am forced to wear a badge [of honor] and municipalities would not allow me to setup business if am an ex-convict.
The same rules never apply to corporates. Municipalities like Mass Turnpike authorities welcome Bechtel with open arms.

Until Senator Leahy introduces a legislation (on the sly) that prevents counties from ALL convicts from establishing a business, this will go on.

Re:It IS a big deal

[Idarubicin]

Say in an "old school voting system" you had a company in charge of transporting ballot boxes from the booths to the counting stations, and one of the trucks took a "minor detour" on the way, maybe for the convenience of the company or the employees (take a leak or buy a drink etc).

I can't speak for other jurisdictions, but I know in the "old school" country of Canada (with its valuable dollar, its gay weddings, its high-quality ganja, its proven oil reserves second only to those of Saudi Arabia, its polite people, and its rejection of firearms in favour of elegant weapons from a more civilized age) the ballots aren't taken to a central location prior to counting.

Ballots are hand-counted at the polling station immediately after the polls close. The ballots are only physically handled by two individuals (each nominated by a different political party), but representatives from any candidate or party are permitted to observe the process.

Don't be silly...

[Etherwalk]

> Do you want your vote counted by people who can't read a contract?

Don't be silly! The don't have to know how to count. They only need to write code.

in other words...

[Anonymous Coward]

just under 1000 voting machines, essentially a $300 computer, cost $15 million?

I'm in the wrong business...

Welcome to the world of high assurance

[Almahtar]

In high assurance systems, be it equipment on an aircraft (that has obvious life or death impact) or equipment for voting (that has indirect, but just as serious consequences), any change has to be certified.
 
There are lots of ways a seemingly ineffectual change can cause unforeseen consequences, so it's not up to the person making the changes to decide what is and isn't a 'trivial' change. Even if the changes can now be proven to be trivial and no danger to proper operation, this is still the right move to make -- people can't be allowed to bypass the controls they agreed should be put in place. I hope this doesn't do them under, because I can sympathize even if I think voting machines are a bad idea. I'd love to see them gone, I'd just like it to be done for a reason that proves someone else shouldn't come back and try to make a similar product but follow the rules.

Elections in America are COMPLICATED

[l2718]

I see absolutely no downside about doing it the traditional way. Is there any reason to do it with machines in america, or you do it that way just because it's cooler?

The US, being an enormous country, has a many levels of government. Unlike many other countries, it runs all elections for all levels of government on a fixed date (some Tuesday in November), rather than spreading them around the year. Of course, not every position is up for election every year, but still this means that the "ballot" contains tens if not hundreds of separate elections, ranging all the way from the US President to the county water board and the town mayor, not to mention multiple "ballot initiative" (direct legislation). Each election (especially president, governor etc) can feature tens of candidates (most of them irrelevant). Printed ballots are thick booklets; both filling them correctly and manually reading them is a non-trivial operation. Also, manually tallying the votes in these hundreds of elections takes a lot of time.

This is not to say that this was not done manually in the past, but certainly using computers greatly simplifies the process. I think the best solution is to use computers to generate the ballot, but only use computer counts provisionally. That is, the voter will step up to a computer and will make selections, after which the computer will print a filled ballot that can be optically scanned. The computer will also tally the votes giving a quick result for most of the races. Nevertheless, the printed ballots should be considered the official votes, the ones to be used if a recount is necessary. In important races (President, Governor) it's probably better to automatically count the printed ballots and only use the computer counts for provisional results. Note that this also allows for people to manaually fill their ballots if they feel like it.

Re:Elections in America are COMPLICATED

[Splab]

So you turned to computers to fix a problem that shouldn't exists in the first place? We also got multiple elections here, not to the granularity of the US but we still manage, how? We hold the elections spread out over the year, makes everything less confusing for the voter so the votes end up in the right place.

Re:Elections in America are COMPLICATED

[Anonymous Coward]

We also have people who live in rural areas, far away from polling places. The more dates you have for elections (and we are coming up on our THIRD this year -- we do local stuff too and then the caucus is Jan 3) the less people you have showing up.

It takes me about 40 minutes to get to my polling place, I don't drive. I don't have child care. Weather is already crappy by election day. I go but why discourage people from going? Once a year is fine.

Re:Elections in America are COMPLICATED

[sumdumass]

From here, you can request an absentee form and either mail it in or drop it off at your convenience up to and including the election day. seems like that would solve the problem for people like you while not effect others.

Re:Elections in America are COMPLICATED

[pig_man1899]

Luckily we have a strong case of voter apathy in America to keep the total number of ballots that need to be counted to a reasonable number. Now imagine if the US had compulsory voting like these countries.

Quite right.

[remmelt]

Oh, good points! So that's why all elections have been failing in the civilised world! Yes, yes, I can see it now, every election in Europe was bought by the largest party, just because they could, it being so trivial!

Or you could have been following the discussion and reading up on the reality of voting and how that works without machines, you would have known that it isn't all as easy and trivial as you like.

Just a couple of pointers to help you on your way: paper trail, possibility of recount, multiple parties involved in each counting hub, economy of scale (if one counting hub somehow gets corrupted, it's just that hub, not the entire state as it is with machines), rooting of voting machines trivially easy as has been shown by people who installed a chess program on it, closed source code in the machines, etc.

So yeah, great one.

Re:Quite right.

[AvitarX]

In Europe minority votes count for something and you have more than 2 credible parties. That makes rigging an election far less valuable.

In the US just a few thousand votes in a key are brings your party from a lot of control (president) to very little (Democrats in the Senate and House can't get shit done).

With razor thin margins and 49% of the vote counting for nothing it is possible to subtly change the votes and drastically alter the political landscape.

In 2000 Florida was withing 1/100th of 1% (0.01%) and would have made Gore the president. I doubt there is many places in the rest of the world where so few people in such a small area could cause such a dramatic shift. This makes the risk/reward analysis in the US much different than other places with your minority reresentation and 3+ parties.

Re:Quite right.

[Anonymous Coward]

Republican President and Democrats in the Senate and House can't get anything done
conversely
Democrat President and Republicans in the Senate and House can't get anything done
Not being able to do anything can be considered a good thing.

Re:Quite right.

[Dirtside]

In Europe minority votes count for something and you have more than 2 credible parties

As a U.S. citizen, I'd be happy with more than zero credible parties.

Re:Quite right.

[AvitarX]

Yes, but when it comes down to it 1/100th of a percent of the voting population in a single state was the difference between our current deadlock, and possible sweeping reforms.

I am not saying either is good, but a very small amount of votes amplified signifigantly the power balance, and I think that is a fairly unique setup (for good or for bad)

Re:Quite right.

[VGPowerlord]

Are you saying that the Libertarian, Green, or Constitution parties aren't credible? ;)

Re:Quite right.

[Dirtside]

Mostly what I object to is our two-party system. If any party besides the Democratic or Republican Parties ever became relevant to U.S. politics on a large scale, they'd inevitably end up just as bad as those two parties are now. Oh, it'd take a while, but it'd still happen.

As long as our electoral system works the way it does now, all political parties in the U.S. will either be massively corrupt or totally irrelevant.

Legal Defences and Contributory Negligence

[Anonymous Coward]

1) The state has a duty of care to SAMPLE and test machines on an ongoing basis.
2) Certification. Fancy word, and may not mean what you think - have to read the fine print.
3) Evaluation negligence. If the MK1 machines were flawed, and MK2 less flawed, no damage took place- remember NT 4 was certified as 'secure'.
4) nAs the state is not posting the 'contract' online, one presumes their purchasing area is in a weak position
5) In Australia, One state decided NOT to pay ongoing maintenance fees after purchase. Turned out the contract said they 'lost certification' after so long, if programmed maintenance from vendor Y was not continued. Thus they did not get compensation as the machines had already lost certification.
6) These days, the profit is in the maintenance contract - - good chance someone has not read the fine print.

No, stealing elections is complicated

[Anonymous Coward]

In our large, Democrat controlled cities, the vote totals in certain wards are run up to bias the statewide totals for Senator, Representative, Governor, and especially President. Can't do that with a clear box and paper ballots. With electronic voting more ballots for the Democrat are cast than total voters.

Kennedy and Clinton (both times) won this way.

Look at city-data.com. Michigan has a monopoly on the 10 counties that went heaviest for Kerry. Many were 90%+ for Kerry. Think that's an accident?

Re:No, stealing elections is complicated

[Anonymous Coward]

Michigan has a monopoly on the 10 counties that went heaviest for Kerry. Many were 90%+ for Kerry. Think that's an accident?

Put your tinfoil hat away. Michigan counties generally use those scantron-type ballots with the black marker-pens, not Diebold or ES&S black boxes. While the sparsely populated west side of the state tends to be Republican, eastern Michigan tends to be heavily Democrat with the big cities (Ann Arbor, Detroit), all the college suburbs and the various labor unions.

I'd be more worried about those Ohio counties where they returned more votes than actual residents.

What state are YOU from?

[zippthorne]

I have voted in two states (in different elections, although disturbingly there doesn't seem to be an effective mechanism in place to prevent simultaneous), and in both the ballot was a single, sometimes double-sided piece of large card-stock. Bigger than legal-sized paper, but certainly not so big as a booklet.

I agree with your second paragraph, except I don't want to waste time and money on electronic voting for everybody. It should be a couple of machines for people with special needs. I can wipe a marker across a scantron bubble as fast as I can push a button. Why should I have to wait for something to print out after that?

Re:What state are YOU from?

[Discordantus]

Not sure about the GP, but I'm from California. The last election I voted in had (iirc) four sheets of paper as the ballot. First came national elections (president, senate, house), then state elections (governor, state house and senate), then local (sheriff, mayor, city council), then eight or nine propositions. This easily covers a few sheets of cardstock.

Re:What state are YOU from?

[mOdQuArK!]

I can wipe a marker across a scantron bubble as fast as I can push a button.

Any procedure relying on manual marking is going to be problematic - you'll get people putting little dots that don't register, big Xs that cross multiple dots, crossing out choices they didn't mean to make & filling in new ones, doodling around the select dots, etc.

A printing machine can create a nice clean user & machine-readable ballot which contains ONLY the choices that the user wanted to make, with no chance of mistaking those choices for anything else. Printing machines can provide a fair amount of benefit.

The same can't be said for the counting machines, however.

Re:What state are YOU from?

[Pentavirate]

I've voted in 4 different states inclucing California and California is a bit unique. They pioneered the direct voting process (initiatives) and they use it more than just about any other state. Most of that ballot consists of the initiatives up for vote. Most states are lucky if they have 3 or 4 initiatives on any given ballot.

Personally I prefer the way Oregon does it. Every registered voter votes absentee. It comes in the mail, you connect the bars with your black pen, then they run it through a scantron like machine. I'd support every state adopting this.

Re:What state are YOU from?

[Pentavirate]

It weeds out the votes of people that aren't smart enough to be voting in the first place (only kinda joking).

At one time it was thought that only property holders should be allowed to vote because they were the ones that actually had a stake in government. I don't agree with that but I do get frustrated when people don't really take a look at the issues and vote for the person that will give them more. I don't mind people voting for a position that I don't agree with as long as they understand the issue and believe in how they vote.

Re:What state are YOU from?

[Discordantus]

It would appear that this scheme would also effectively eliminate the exit poll, which would be (said Martha) A Good Thing.

However, I wonder how this impacts people who have no stable postal address?

Re:What state are YOU from?

[Pentavirate]

True that they couldn't do exit polls but the ballots have to be in by election day so they can run all the ballots that have come in up to that day and get a good idea of what the trend is. It's got to be at least as accurate as exit polls.

As for stable addresses, I'm not sure how they handle that. I believe that you can still go down to the county building and pick up your ballot and turn it in there if yours never arrives in the mail, so if your address is unstable, you'd have to do it that way.

I absolutely love absentee voting. I first signed up for permenant absentee status in California and it was invaluable to be able to sit down with your ballot and your voters guide and spend a few hours one evening and get it all on the ballot as you read the pamphlet. I don't know how Californians can keep it all in their head long enough to get into the booth.

A little off topic, but what's the deal with the voting of a dozen judges you've never heard of with scant information in the voter's pamphlet. I usually end up not voting for any because I don't feel like I can make an informed decision.

Re:What state are YOU from?

[riverat1]

In Oregon if you are homeless but want to vote you can have the ballot delivered to a homeless shelter or other service agency. Also, you can go to the county elections office and fill out your ballot there. Since the ballots are mailed to voters about 2 weeks before election day you can do that at any time during that 2 week period.

I have to say I'm happy with this system. It give you time to consider the ballot and take your time filling it out. I do suspect that in some authoritarian families there may be some people forced to vote for the authority figure's choices but I've never heard of a problem with this so I doubt it happens a lot.

Seduced by technology

[bflynn]

America and Americans are generally seduced by technology. It is in our culture to invent and to innovate. We're always pushing for the next thing, always trying to go further. If you need any proof, just look at our Navy - all nuclear, top of the line technology. And yet, an inexpensive Chinese submarine can appear in the middle of a battle group.

At times, this goes too far. Voting is one of those areas. There is nothing accomplished by computerized voting machines that could not be accomplished with a piece of paper, a black marker and some volunteers to count. After all, tallying a vote is just that, counting. And it is infinitely more difficult to tamper with a properly controlled paper balloting system than it is with a computer voting system. Perhaps it is "old-fashioned", but I see nothing wrong with marking a piece of paper.

Re:Seduced by technology

[TimTheFoolMan]

That is, as long as you're not physically unable to mark a piece of paper by yourself.

For all its faults (and there are many), the Help America Vote Act (HAVA) came about because people who are blind or physically unable to mark a ballot had no way of voting independently and privately. To that end, their civil rights were not being addressed by the individual states, and the resulting legislation forced the states to come into compliance (well, everyone except New York).

Prior to the DRE (Direct Recording Electronic) variants, we did all manner of technology things to try to eliminate the nefarious things that various groups would do to unfairly influence elections. Lever machines, scantron/marksense systems, and so on, were all attempts to get the "personal touch" that was so frequently applied, removed from the process. Lever machines seem to have had the best reputation, but even those were susceptible to tampering, in subtle (and frequently invisible) ways.

Another issue in the US is the complex nature of voting rights. Voting laws and regulations are the province of each state, even when it comes to selecting the Electoral College representation for Presidential elections. For example, KY (where I live) could decide to choose its electors by flipping a coin, and our friends in IN or TN couldn't do a dang thing about it. As long as the states don't do anything that biases the process in favor of one particular group over another (such as male/female, white/non-white, non-disabled/disabled), the feds have no say in the matter.

Lastly, you have the issue of US geography. There are many places in the US that are incredibly rural, where outsiders are simply unwelcome. Smart people, even federal agents, go into the hills of Eastern KY with caution, because they know that going in and throwing their weight around so carries a fair amount of risk. I know a former FBI agent who NEVER traveled into Eastern KY alone for just this reason. I would expect that each state has areas like that, where outside review of voting practices or oversight will not be received gladly. In those areas, the states are always looking for ways to get family/regional influence out of the process, because bipartisan oversight is so incredibly laughable.

Like most issues on Slashdot, this is a lot more complicated than the average person (especially those outside the US, or with little familiarity with the US version of representative democracy) realizes. While I see no reason to cut ES&S any slack on this matter, I can sympathize with the difficulty of navigating the plethora of regulations and laws that such a company is subject to, should they choose to sell to more than one state in the US.

Tim

Arguments

[Shaltenn]

From us people on /., we know that a moved circuit board and a different bracket has little chance to make a real difference with the operation of a machine.

However! To the average person, they look at item A, then look at item B, and see that something is not in the same place and say "These must not operate in the same function - which isn't all that unreasonable the more you think about it. The issue remains clear: They have two different machines (at least structurally) - and they must be certified before sale. They were not certified, and thus got smacked down. That's the way it works.

Good for california.

Don't forget....

[foniksonik]

If CA had allowed them to do this, without bringing suit, then the CA gov would be liable if there were ANY problems - real or imaginary, which could somehow, anyhow be traced back to this discrepancy. The State Sec or State is doing the right thing both for the people and for his job security.

A court may find that the damages are too great, who cares... he brought the suit and is now off the hook for anything that may or may not have happened come election time.

Engineering controls

[TheHawke]

ES&S apparently does not have a sufficient engineering and/or QA certification system in place or they would not have pulled this expensive stunt.

Simply relocating boards and wiring would constitute a simple modification submission to the right authorities with proof that it would NOT affect system performance in any way, shape or form. Sure, it's lot of paperwork, but consider the consequences.

This reeks of their marketing pulling a stunt to make more money for the company.

10k penalty plus cost of the machines?

[someone1234]

They wanted the machines for free:)

I think you're wrong.

[wonkavader]

Printing ballots has a much higher profit than selling voting machines. If you sell a voting machine, you make a bunch of money, right off, but there's R and D, manufacturing costs, etc.

Printing has a real margin, and the revenue keeps recurring.

The most profitable thing IBM ever made, up until the IBM PC, was their PUNCH CARD.

I think support for the machines might generate enough recurring revenue to beat the paper, but the machines themselves are probably not as good a way of taking money from a government.

Unless, of course, you can get them to buy a voting machine, then chuck it and buy a new one every, say, three years. There's no reason for a properly designed voting machine to be chucked, though. This is probably part of why there's so much work being done to make crappy machines. You want them to be a laughingstock in a few years, so you can get recurring, not one time, revenue.

Re:I think you're wrong.

[Anonymous Coward]

I recently read a book about the 2004 elections. These machines are so rigged it's not funny.
In Vegas the Nevada Gambling commission requires all machines to be certified, and the source code placed in escrow. When someone tried to get the voting machine makers to provide source code (in order to verify election results) they claimed trade secrets.

In Canada they have a NON PARTISAN commission in charge of Federal and a 2nd one in charge (in each province) of provincial elections.
The problem with America is not so much rigged machines or games that the various parties play (thought that is a big part) is that the elections are Partisan and are subject to the whim of some party hack in each county.

You can have a different voting system in each county (thus the butterfly ballot, pregnant and/or hanging chads... Its a mess and it will NEVER be fixed till it comes non partisan and is not controlled by the government of the day

It's even worse than that...

[LynnwoodRooster]

See, here we don't actually CHECK (by law) if you're a citizen or not, when you register to vote. When you sign up for your drivers license - citizen or not - you're given the opportunity to register to vote, so anyone with a US driver's license (which only requires proof of address and identification - a credit card and a cell phone bill will suffice for those two items) can register to vote.

And in fact, by LAW we're prohibited from even checking to see if you are who you say you are when you go to the polls to vote. Because after all, requiring people to bring an identification card with them when they are going to vote is "racist" and "totalitarian"...

Heck, you can just show up at a polling place, say that you're supposed to vote EVEN THOUGH YOUR NAME IS NOT ON THE LIST, and be handed a provisional ballot that you can fill out, and have a good chance of having it counted. Never actually registering OR showing proof of identification.

And now we're actually MAILING the ballots to you, so you can fill them out and mail them back. Never mind that you may not get a ballot, or can register as many times as you like and fill them all out. Or take your neighbor's ballot out of their mail box and vote for them...

The "problem" with American elections isn't the machine used, or butterfly ballots, or "evil corporations" like Diebold. The problem is that we have ZERO control over who becomes a registered voter, and who actually votes. Until that problem is solved all the huffing and puffing about machines being used to steal votes is worthless. It's a LOT easier to steal an election if you cannot control the voter rolls to start.

We need to wipe clean ALL voter rolls, then rebuild them only with names of people who can prove they are legally allowed to vote - citizens. And then when you DO vote, you have to show up at an actual polling place and prove you are who you say you are.

Otherwise, pencils, pens, electronic, pull levers - none of the actual ways votes are cast - matter. The most critical part of voting begins with the voter rolls, not with the actual casting of the ballot.

For the record, I am a US citizen, registered in the state of Washington (where I live when I am in the US), and we have an on-going problem in the biggest county in the state (King County) where it's quite common to have more votes than voters (voters being people who signed in at the polling place), and more voters than registered voters. Or to have a lot more voters than actual votes (meaning lost votes), depending upon the political leaning of your particular district. Oh, and don't forget that we have ballots "found" weeks after the election, yet still counted...

Cleaning the voter rolls and controlling access to the ballot box are the two most fundamental steps towards a secure election. Worrying about what the ballot box actually IS - that's just a diversion to make the public think something's being done about the elections messes...

Re:Offtopic?

[Anonymous Coward]

What do you mean, "offtopic"? Ponies are ALWAYS relevant.

Don't forget Washington and Minnisota

[HornWumpus]

Those were successful Democrat fixes in '04. You don't want to talk about those? What a surprise. Better evidence of crooked voting there then Ohio or Floriduh.

Both major parties are big time cheaters and have been sense way before electronic voting machines. That's why nobody with any power wants a real investigation.

Re:Don't forget Washington and Minnisota

[doom]

Those were successful Democrat fixes in '04. You don't want to talk about those? What a surprise.

Why wouldn't we want to talk about them? Is this the old "but democrats do it to!" defense? Even if that were true, it's hardly reassuring.

Better evidence of crooked voting there then Ohio or Floriduh.

Oh bullshit. I was following the Washington story and it hardly deserved to be called a "fix". It was however, entertaining to watch the Republicans get nailed in a close election ala Florida, and take a completely different line when it was their ass getting screwed.

They have absolutely no shame. It's amazing what the media let's them get away with.

Both major parties are big time cheaters and have been sense way before electronic voting machines. That's why nobody with any power wants a real investigation.

"Well you see, we're not going to talk the slashdot crowd into voting Republican, but they're a paranoid bunch, and if you play up that you can make them apathetic about going out and voting Democrat..."; "Okay, that sounds good. Play it that way."

Any bets on how this post is modded?

[HornWumpus]

I'm guessing flamebait not informative.

Voting machines have nothing to do with reg fraud.

[HornWumpus]

The D's traditional method of rigging votes.

See also St. Louis county Missouri and King county Washington for truly egregious examples of registration fraud and votes being delivered to order late in the day when the D's are about to go down.

Not exactly the same

[Anonymous Coward]

In the state suit, it is stated that the AutoMark A100 was suppiled with and certified for version 1.0 of the firmware, and that the A200 ships with version 1.1.2258 (section V "Facts", paragraph 26). Does anyone reading /. really think that v 1.0 is exactly the same as v 1.1.2258?

Despite the summary, and the company claims, if true this is not just the relocation of some system boards and a different paint job.

See the magic machine, and why it will never work

[Catbeller]

Voting machines such as those ESS and Diebold sell are useless for anything but cheating... as this illustrates, you can validate code and approve the circuitry design all you like, BUT YOU DON'T KNOW WHAT IS ACTUALLY IN THE MACHINES AT GO-TIME. They can place cheat chips or upload cheat code, and by the time anyone even has an inkling beyond simple statistical obviousness -- which American media apparently can't understand -- the election is years over and we are told no one cares. Unless of course a Democrat wins; imagine the hell unleashed by FoxNews! if a single voting system may have scooted a Demo into office with some question of the system's honesty. But I digress.

Paper trails are useless in the usual conception as well. A voter votes, and a little piece of paper comes out telling him that he voted thusly. Oh, PLEASE. Unless the paper is a card and can be manually recounted and the totals compared to the computer tally, the paper trail is worthless. Anything can be recorded, and anything can be printed. Now, if instead for security purposes the paper card receipt is kept archivally, then why the HELL have a PC acting as an agent in the first place, and simply count the paper cards? Canada uses a number 2 pencil and paper ballet, and they count the votes manually, with a rep from both or more parties observing the count-- you know, the thing that the Supreme Court here said didn't work. And they finish national elections in hours after the polls close. Faster than we do. Their method scales, you see.

There is no reason to computerize the voting process other than cheating. None. All else is sophistry. We had working systems; they were abused by injecting doubt in Florida in 2000. The recounts work fine if the lawyers and the Supremes stay the hell out of things. None of you may recall, but in 2000, at the same time Florida was being sued and stayed to death and back, two more recounts were happening out in the western states -- manually -- and no one said a thing. Florida was made a carnival by Republicans because they wanted to instill the idea in a fantastically compliant media that recounts didn't work, that chads made things uncountable somehow, that NO RECOUNT WAS NECESSARY by any means possible. It took faked up riots in Dade by republican staffers pretending to be random thugs demanding a shutdown and a crooked Supreme Court majority -- all rightist Republicans, and I include Kennedy as he has shown his new colors since then -- to order the shutdown of the democracy hours before the recount was supposed to end. Never has the US seen a group of election officials and volunteers work so hard and so quickly to beat a crooked shutdown and what was frankly a putsch by the Republican party.

How different the world would be now if Gore had been allowed to win. The worst. Day. Ever. In American history.

"evotes rigged 2004" is not a crazy notion.

[doom]

what an intriguing mix of insight and paranoid nonsense.

Voting machines do not exist for people to buy elections. They exist because it monetizes the election process, allowing people to get wealthy by controlling a process that is required in a democracy.

You're jumping to the conclusion that because you understand one aspect of the situation, you understand all of them. There's some statistical evidence that there were some funny things going on with electronic voting machines in the 2004 election: Who won?.

The notion that the Republicans stole one (or two?) presidential elections in a row might be wrong, but doesn't deserve to be dismissed as "paranoid nonsense". At the very least, you'd think it would be a high priority to make sure that such things aren't remotely possible in the future...

stolen elections and proof

[doom]

"They"? Who are "they" and what election did they steal? Proof? Something from a reliable source, please?

Listen: have you looked into this at all? Why is it the problem of some-guy-on-slashdot to bring you up to speed on what ought to be common knowledge at this point? In summary, there have been three different styles of attack in play in US elections (1) denial of the right to vote for people likely to vote against you; (2) shorting key districts of voting machines; (3) falsify the vote itself, via electronic voting machines. Oh, and you might throw in a number (4) manipulation of the government legal system to smear the opposition.

If you're in the mood to look up supporting data on each of those three points, for (1) I might suggest reading Greg Palast on the subject, (2) try reading some Ohio Free Press articles (they're online), also that HBO documentary had some striking footage of the problem (3) I suggest reading Freeman and Bliefuss on the subject (4) Is how Alberto Gonzales got chased out of the Attorney Generals office, it's not exactly obscure.

My apologies for not doing the link farming for you, but I'm getting tired of playing co-dependant with the willfully ignorant.

If you ask me, there's certainly been enough proof of chicanery to justify an investigation into the problem, but that hasn't happened. And there's definitely, shall we say, "cause for concern" about the integrity of American elections. Debra Bowen's election to Secretary of State of California has been one of the few bright spots in recent years.

economic motivations ain't everything

[doom]

Opportunist wrote:

It's not like the vendor of the voting machines would want to rig an election. What for? They'd get paid by either party anyway, so why bother?

And here we see the problems with working from economic theory instead of actually looking at what's been happening. People do all sorts of things for all sorts of reasons, not all of them are economically motiviated. The folks at Diebold famously had a strong Republican bias, and there are strong personal connections between the people running these electronic voting machines companies (to quote Freeman and Bleifuss again: "The voting machine company Datamark, which became American Information Systems and is now known as ES&S, was founded in 1980 by two brothers, Bob and Todd Urosevich. Today, Todd is a vice president at ES&S and Bob is CEO of Diebold Election Systems.").

You want me to explain why someone would want to do what they do? What, am I a mind reader? If you talk to Mark Crispin Miller on the subject, he'll start ranting about Republican "theocrats". Maybe he's right, maybe not, how are we supposed to know?

The obvious point should be that we need a simple transparent election process to make sure that such things don't happen, even if no one is convinced (or wants to believe?) that they already have happened.

When you have a paper-only voting process, simply dump the votes on the protester and tell him to recount.

Actually in San Francisco, there's a history of dumping the votes in the Bay to make sure that they can't be counted at all, but that's another problem.